Mustafa Berke Yelten

A cryogenic modeling methodology of MOSFET I-V characteristics in BSIM3

Transistor models for circuit analysis have temperature dependent parameters which are only valid in standard temperature range of −55 °C to 125 °C in general. Circuits designed for low temperature military and space applications should work in cryogenic conditions properly. Thus, transistor models must be modified. In this paper, a methodology has been developed to optimize these parameters for MOSFET devices at low temperatures. The methodology updates all parameters regulating the temperature dependency of the drain current, threshold voltage and saturation velocity based on the chosen target data set taken at low temperature. An automated system involving a circuit simulator and mathematical programming tool has been established that can analytically compute revised model parameters independent of the transistor process. Using this methodology, average errors in I-V curves of various sizes NMOS and PMOS transistors for 0.18 μm technology has been reduced below 2.5% and 3.5%, respectively.

Efficient signature selection tool for sense & react systems

Reconfigurable circuit design has become very important in the last decade for increasing the lifetime of CMOS circuits in deep sub-micron technologies. Sense & React approach is one of the reconfigurable design approaches, where degradation in a circuit performance is detected via sensor circuitry and a pre-established recovery operation is applied to circuit. However, sense operations are quite problematic since direct measurement of the degradation in circuit performance is highly complicated. Therefore, indirect measurements are preferred, in which one or more relevant circuit variables, which are called signatures, are selected out of measurable circuit quantities. Conventionally, the designer selects the signatures by performing an iterative search and evaluation on aging simulation results, and no such procedure has been defined in the literature yet. This paper proposes an efficient selection methodology and tool for determining efficient signatures for sense operation.

Review: Analog design methodologies for reliability in nanoscale CMOS circuits

In modern CMOS technology, local electrical stress has substantially increased as device geometries scaled down more aggressively compared to supply voltages. As a result, time-dependent degradation mechanisms (aging phenomena) became an important performance problem, which leads to a considerable lifetime reduction in manufactured integrated circuits. Combination of these aging phenomena with process variations has made reliability a major design objective. In analog circuits, different approaches have been proposed to mitigate the performance challenges related to device reliability. This paper discusses aging in CMOS technology and reviews reliability-aware analog circuit design methodologies for nanoscale circuits.

Reliability of 3D-printed dynamic scanners

3D-printed dynamic structures have arisen as a lower cost and easier to fabricate alternative to miniaturized sensor and actuator technologies. Here, we investigate the reliability of a selected 3D-printed laser scanner, which was initially designed for miniaturized confocal imaging, having 1 x 1 cm2 footprint. The scan-line, 1st resonant frequency and quality factor of 3 devices were monitored for 100,000,000 (hundred million) cycles, and an average deviation of <6% was observed for all three parameters under investigation, for the devices under test. We conclude that 3D printed dynamic structures are promising candidates for a variety of applications, including optomedical imaging applications that demand disposable and low-cost scanning technologies.

A heuristic sensitivity analysis technique for high-dimensional systems

A viable technique for sensitivity analysis in high-dimensional systems is described in the context of bio-inspired systems. The sensitivity analysis provides critical information about the system by indicating the dominant parameters that shape the output. This knowledge becomes particularly essential to have a better understanding of complex biological network of interactions. A notable feature of many high-dimensional systems is that a large portion of all parameters have little impact on the system outcome, thus yielding sparsity. The proposed algorithm leverages the sparse properties of systems analysed and is based on a heuristic two-stage elimination strategy. The implementation of the proposed algorithm yields substantial reduction of the total simulation cost by as much as 95% for a system composed of 562500 parameters over the conventional local sensitivity analysis while retaining its accuracy above 70%.